Predicting corporate credit ratings using the content of ESG reports
摘要
Environmental, social and governance (ESG) reporting has emerged as a critical element in assessing a company's sustainability and ethical impact, influencing its financial performance and investment attractiveness. Despite its growing importance, there remains a gap in understanding the direct link between ESG practices and corporate credit ratings. To address this issue, our study proposes a novel approach by analysing the content of internal and external ESG reports to predict credit ratings. Data from 60 large German companies for the period 2019–2022 are analysed using a fine-tuned MPNet language model to reveal the ESG topics, along with the sentiment and subjectivity of the reports. Our findings show that incorporating ESG factors from the reports leads to a more accurate prediction of corporate credit ratings, highlighting the value of ESG considerations in financial assessments and their broader implications for investors, policymakers and corporate strategy.